The major objectives of the proposed research are to examine influences on the sexual and contraceptive behavior of two groups of women at high risk of contracting AIDS--partners of intravenous drug user and prostitutes; and to identify variables that predict safe sex practices and the decision to have HIV testing. Specific goals include examining the relationship between sexual and contraceptive behaviors and: (1) variables important in the adoption of health behaviors (e.g., knowledge of the disease and perceived vulnerability to the disease); (2) self- efficacy regarding sexual and contraceptive behavior; (3) empowerment; and (4) demographic characteristics such as ethnicity, age, and parity. Other specific aims are to describe retrospectively reported changes in sexual and contraceptive behaviors in response to fear of AIDS; to compare the sexual and contraceptive behaviors, as well as other characteristics, of women ever diagnosed or never diagnosed as having an STD (including AIDS or ARC, genorrhea, genital herpes, and chlamdydia); and to examine differences between high risk women who decide to have HIV testing and those who do not. A total of 1500 women who are prostitutes or partners of intravenous drug users will take part in personal interviews in three cities, Los Angeles, Phoenix, and Boston. In each city respondents will be reached through public health department clinics; drug treatment and prevention programs; the criminal justice system; AFDC, food stamp an unemployment compensation centers; and an appropriate community setting for each city, such as housing projects, churches, or organizations for prostitutes. The interview will cover parity, alcohol use, depression, personal efficacy, empowerment, peer group and partner support for safe sexual practices, knowledge and beliefs about AIDS, changes in contraceptive use, condom use, and HIV testing intentions. The results of this research should be valuable in the promotion of safe sexual practices among the target populations. In addition, they might be useful in designing specific types of interventions for particular subgroups (e.g., adolescent prostitutes).